1. Field of the Invention
The present invention relates to an organic electroluminescent device (hereinafter, abbreviated as an “organic EL device” or “device”) which can be used in planar light sources and display devices.
2. Description of the Related Art
Recently, attention has been drawn to organic electroluminescent devices having a light-emitting or luminescent layer including an organic compound between a cathode electrode and an anode electrode opposed to the cathode electrode as a large area display device operable at a low driving voltage. For the purpose of higher efficiency in an EL device, Tang et al., as is disclosed in Appl. Phys. Lett., 51, 913 (1987), have successfully achieved a sufficiently high luminance and high efficiency for practice use, i.e., a luminance of 1,000 cd/m2 and an external quantum efficiency of 1% at an applied voltage no more (greater) than 10 volts, by adopting a structure in which organic compound layers having different carrier transporting properties are laminated to thereby introduce holes and electrons with good balance from an anode and a cathode, respectively, and by having the thickness of the organic compound layer no more (greater) than 2,000 Å.
Furthermore, according to the disclosures of the patents invented by Tang et al., (such as Japanese Laid-open Patent Application Nos. 59-194393, 63-264692 and 2-15595 and U.S. Pat. Nos. 4,539,507, 4,769,292 and 4,885,211) it is stated that if a total layer thickness of the organic layers sandwiched between an anode and a cathode does not exceed about 1 μm, an EL device capable of emitting light at a lower level of the applied voltage can be provided, and that desirably, if the total layer thickness is reduced to a range of 1,000 to 5,000 Å, an electric field (V/cm) useful in obtaining a light emission at an applied voltage no more than about 25 volts can be obtained.
The reason why Tang et al. have directed their attention to a reduction of the layer thickness of the organic layers in attaining a reduction of the driving voltage, as described in the above-referenced article, resides in overcoming the problem suggested by Helfrich et al. in the 1960s. Namely, Helfrich et al. have observed that an external quantum efficiency of about 5% can be obtained when a sufficient electric field electroluminescence (EL) is applied to an anthracene single crystal; however, according to their method, only a low power conversion efficiency (w/w) could be obtained, since the voltage required to drive such devices is quite high (greater than 100V).
Referring to the above-reference Tang et al. patents, the organic EL devices suggested therein have a multilayered structure in which an anode, a hole injection (transporting) layer, a light-emitting layer (having an electron transporting property) and a cathode are laminated in that order, and the devices can provide a quantum efficiency of at least about 5×10−4 (0.05%). Furthermore, the quantum efficiency is defined in Japanese Laid-open Patent Application No. 59-194393 as the EL quantum efficiency simply equaling the ratio of photons per second emitted from the cell, to the electrons per second measured in the external circuit.
Presently, as has been already disclosed, when a fluorescent material (utilizing emission from a singlet excitation state) is used in the thin layer EL devices suggested by Tang et al., a quantum efficiency above 5% can be obtained. Furthermore, when a phosphorescent material (utilizing emission from a triplet excitation state) is used in the EL devices, a quantum efficiency approaching to 20% can be obtained.
As can be appreciated from the above description, the quantum efficiency is calculated from the number of the photons actually emitted from (outside of) the device, and thus the quantum efficiency is called external quantum efficiency. On the other hand, the number of photons generated internally in the device might be quite large when compared with the value observed externally, and it is predicted that such efficiency, called internal quantum efficiency might reach about 5 times of the external quantum efficiency. Accordingly, even presently, when using a phosphorescent material, an internal quantum efficiency can be exhibited at 100%, and thus it seems that the remaining problem in the organic EL devices resides only in an increase of the reliability concerning the operational life-time of the devices.
As described above, the suggestions by Tang et al. in their patents and articles have accelerated a worldwide research and development in the field of organic EL devices, and thus a great number of improved EL devices have been developed based on the basic device structure suggested by Tang et al. Presently, commercialization of the EL devices has already started in regard to their use as a display device on a dashboard or in a cellular phone.
However, from a viewpoint of durability of the device, the above-described conventional organic EL devices can barely attain a half-decay life time exceeding 10,000 hours with a luminance of only the order of 100 cd/m2, which is required in display use. Presently, it is still difficult to attain a required practical operational life-time (10,000 hours or more) with a luminance of about 1,000 to 10,000 cd/m2, which is required in illumination use, etc. In fact, an organic EL device having a high luminance and long operational life-time is still not realized and not commercially available.
As described above, attention that has recently been drawn to organic EL devices has been based on the discovery of a thin film-forming material which drives the resulting device at a low voltage of not more than 10 volts. However, the resulting device still suffers from the disadvantage that if the device is intended to obtain a high luminance emission necessary for illumination purposes, a higher current density approaching tens of mA/cm2 to hundreds of mA/cm2 is necessary. Note that in the best green light-emitting devices currently available, a luminance of about thousands to tens of thousands of cd/m2 still needs the above-mentioned current density of about 10 to 100 mA/cm2. It can be considered that this property is characteristic of charge injection type devices (like this organic EL device), and such characteristics can cause a relatively large problem with the operational life-time of organic EL devices in comparison with an inorganic LED (light-emitting diode) which is also a charge injection device and uses an inorganic compound semiconductor which can be more robust than organic compounds.
In an organic layer formed from a low molecular organic material via a vacuum vapor deposition method, the nature of the electric current passing through the organic layer is defined as a hopping conduction of electrons and holes between the molecules of the material. Furthermore, when observing the molecules from the chemical aspect, it can be described like this; the electron transporting molecules and the hole transporting molecules which are generally being as electrically neutral molecules are repeatedly subjected to a process in which the electron transporting and hole transporting molecules are shifted to a radical anion state or a radical cation state, i.e., the oxidation-reduction reaction in terms of Lewis' chemistry is being repeated between these molecules. Referring to the above-described property in the organic EL devices, i.e., that a higher current density is required to attain higher luminance, this property means that the oxidation-reduction reactions are repeated at a higher frequency. Obviously, the deterioration speed of the organic molecules is proportional to a frequency of the oxidation-reduction reactions, namely, the current density.
To solve the above problem, Japanese Laid-open Patent Application No. 11-329748 (corresponding U.S. Pat. No. 6,107,734) suggests an organic EL device in which a plurality of organic light-emitting layers are electrically connected in series through an intermediate conductive layer, and with regard to the intermediate conductive layer, describes that many types of materials may be used in the formation of the intermediate conductive layer, as long as they (the intermediate conductive layer) are capable of injecting holes and electrons to one or the other primary surface side, and capable of keeping an approximate equipotential in the layer.
This EL device, however, suffers from the following problem. For instance, in the display device having a simple matrix structure, the light emission area upon voltage application should be defined only to the pixel, i.e., the intersection area, sandwiched by cathode and anode line, thereby enabling to display a motion picture. However, in the above-described case in which the intermediate conductive layer having a substantially equipotential surface is formed in a substantially overall surface in an area which is equal to the area of the organic light-emitting layers, i.e., when the intermediate conductive layer is also formed in areas other than the intersection area sandwiched by cathode and anode line, light emission can be generated in areas other than the intersection areas in which the light emission is desired to be generated. Specifically, there is a possibility of generating light emission in all of the crossed area of the cathode with the intermediate conductive layer, the crossed area of the anode and with the intermediate conductive layer, and if two or more intermediate conductive layers are contained, the crossed area between one intermediate conductive layer and another intermediate conductive layer.
Accordingly, it is described in Japanese Laid-open Patent Application No. 11-329748 that the intermediate conductive layers of each pixel are separated not only from the intermediate conductive layer of the adjacent pixels, but also from a power source. Furthermore, one idea to separate the intermediate conductive layers from each other in the pixels in the EL device having a simple matrix structure is also described in this publication. If an interlayer insulation film is previously formed and disposed at a layer thickness above 1 μm and in the form of a sharp step pattern, the conductive layer can be automatically separated in the presence of a suddenly-changed shape of the interlayer insulation film, even if the conductive layer is formed using the shadow mask identical to the one for an organic material deposition.
However, in this case, although the cathode should not be separated, the cathode can be separated by the interlayer insulation film if the cathode has only a thickness of about 0.1 μm (100 nm) as in the conventional organic EL devices. To avoid this problem, Japanese Laid-open Patent Application No. 11-329748 teaches use of In (indium) as the cathode material at large thickness, thereby preventing electrical separation of the cathode line, because indium cannot easily cause problems due to crystallization (this problem is generally referred to as “hillock”), even if the cathode is formed at a thickness of 1 μm or more.
In this alternative case, however, a problem of the throughput reduction cannot be also avoided, because a metal such as Al (aluminum), which is a conventional and low-cost wiring material, cannot be used as a cathode material and also it is necessary to stably form “an interlayer insulation film having a layer thickness above 1 μm and a suddenly-changed shape of the interlayer”.
Furthermore, the inventors of the present invention have also proposed another organic EL device in Japanese Patent Application No. 2001-225847, and has at least two light-emitting units constituting the conventional organic EL device (the components in all the elements constituting the conventional organic EL device except for a cathode and an anode), and the contained light-emitting units are separated from each other with a transparent layer acting as an equipotential surface.
The “equipotential surface” used herein means that when a voltage is applied, the transparent layer cannot exhibit a substantial potential difference in both a thickness direction and a planar (lateral) direction in the layer. In other words, although the inventors have not specifically disclosed, they have implied the necessity to construct the equipotential surface from an electrically conductive material, i.e., any material having a resistivity less than 1.0×102 Ωcm.
However, as in the above-discussed Japanese Laid-open Patent Application No. 11-329748, if the two or more light-emitting units are separated using a material having a high electrical conductivity (low resistivity) described above, there may be difficulties in defining light emission areas as required, due to the conductivity in a planar (lateral) direction (direction parallel to a substrate).
In practice, as shown in FIG. 38B, even if the production of the EL device is carried out in accordance with the method of Japanese Laid-open Patent Application No. 11-329748 by producing a cathode 55 and an anode 52, both in the form of a strip having a width of 2 mm, and arranging the cathode 55 and the anode 52 so that they are crossed at right angles, thereby producing a light emission area corresponding to the crossed (intersection) area, i.e., 2 mm square (□), unexpected light emission may be caused in other areas when there is an area having an equipotential surface 54 is extended to another area. The undesirable emission in the EL device is shown in the photograph of FIG. 38A.
To avoid the above problem, as disclosed in the examples of Japanese Patent Application No. 2001-225847, the inventors had to form an equipotential surface using a shadow mask (2 mm square pattern; □) having a patterned opening which corresponds to the desired light emission area, thereby selectively forming the equipotential surface only in the desired emission layer. However, in this method, it is difficult to attain selective emission only in the desired pixels in the display device, because the display device has to be produced at a pixel length and pitch (between each pixel) of about 0.1 mm or less.
In regard to improving productivity in mass-production of the EL devices, frequent changing and precise positioning operations of the shadow mask is not desirable, because it causes tremendous reduction of throughput.